根据从高到低的顺序对Linux文件中的字段进行排序

Question

I have a file in which 3 fields are present.

Field 1 has names, field 2 and 3 has numeric values. I need to sort all the fields based on the highest values present in it.

File1.txt

ASANSOL 16.76   31.991
ASANSOL 16.72   33.687
ASANSOL 16.48   25.344
ASANSOL 16.74   30.777
BANGALORE   13.39   17.243
BANGALORE   14.34   20.852
BANGALORE   15.58   25.112
BANGALORE   14.59   34.752
BANGALORE   17.93   2.792
BHOPAL  16.53   13.487
BHOPAL  16.7    11.599
BHOPAL  16.65   16.003
BHOPAL  16.87   20.674

After sorting and taking unique values, the desired output should produce as below:

ASANSOL     16.76   33.687
BANGALORE   17.93   34.752
BHOPAL      16.87   20.674

Help me how shall I go about this? #Script would be preferable

Answer 1

Sorry I made an answer of not asked first. The right answer is simpler:

awk '{ d1[$1] = $2; d2[$1] = $3 } END { for(x in d1) print x, d1[x], d2[x] }' <filename>

----------- Old answer -----------------

simple answer is:

cat <filename> | awk '{ if($2>$3) { max = $2 } else { max = $3 } print max, $0 }' | sort -r | sed 's/^[0-9.]* //'

1. find max number using awk and prepend it to every line;
2. sort;
3. remove first column using sed

Answer 2

如果您不想编写一个小的脚本来处理此问题，则可以使用awk进行排序： http : //www.gnu.org/software/gawk/manual/gawk.html